Adaptação e validação do teste Profiles of Early Expressive Phonological Skills (PEEPS)

Abstract

Objective: To carry out the cross-cultural adaptation of the Profiles of Early Expressive Phonological Skills test (PEEPS) to Brazilian Portuguese (BP), called PEEPS-BP - Expanded List, seeking validation of content, criteria, sensitivity specificity coefficients and cut-off point. Method: Quantitative, cross-sectional study with psychometric measures, divided into two stages: Stage 1 - Content validation, composed of specialist and non-specialist judges, child judges and a pilot study; Step 2 - Criterion validation, composed of a clinical group (7 children) and a control group (23 children). In Step 1 - Expert and non-expert judges judged 423 words from the MacArthur Inventory, from BP, in terms of familiarity and representativeness. The child judges, 4 participants aged 24 to 36 months, judged the words resulting from the analyses of the expert and non-expert judges, for the application of the PEEPS-BP test – Expanded List. Afterwards, the PEEPS-BP – Expanded List in the Pilot Study with 6 participants was applied to define the data collection, with the necessary modifications in relation to the stimulus words and toys. The children were evaluated to verify if they fulfilled the inclusion criteria, that is, to present typical linguistic and neurological development. In Stage 2, 30 children participated, who were evaluated in order to ensure typical linguistic and neurological development, with 23 children comprising the control group. The clinical group was formed by 7 children with complaints regarding expressive language, presenting neurotypical development. The PEEPS-BP – Expanded List test was also applied in Step 2, in both groups. Regarding data analysis, in Step 1, for the analysis of the results of expert and non-expert judges, the Kappa Fleiss test of agreement and the Gwet test were used. In Step 2, the type of response provided by the child was analysed: spontaneous naming, repetition and refusal (when the child does not speak the word), with different scores for each analysis. In addition, in Step 2, the T Test was applied to the variable of the groups in relation to the total score of the test, considering a significance level p ≤ 0.05. For the sensitivity and specificity coefficients, the ROC- Receiver Operating Characteristic Curve was used. Results: For the results of expert and non-expert judges, the cut-off point, between familiarity and representativeness, scored the words belonging to the extremely familiar and very familiar group, with absolute intersection in representativeness, totalling 122 words. From these, new criteria were established for selection: lexical, phonological and syllabic word structure questions, resulting in 34 words. Regarding the child judges, nine words scored lower than expected, so the toys were changed to the Pilot Study, which ended the PEEPS-BP – Expanded List, with 29 stimulus words, represented by toys. In Step 2, the T Test was applied to the variable of the clinical and control groups in relation to the total score, and the result was significant p<0.001. For the sensitivity and specificity coefficients, the values of the coordinates available in the ROC curve were observed. The positive value scored 34.5 correct answers, with a sensitivity value of 95,7%, with a specificity of 100%. Conclusion: The PEEPS-BP – Expanded List test presents a cross-cultural adaptation for BP, with evidence of content validity, criteria, sensitivity, specificity coefficients and cut-off point that show the difference in the application of the test in children from the clinical and control groups.